Wage inequality and firm growth

By Paul Walker 27/03/2015

At VoxEU.org Holger Mueller, Paige Ouimet and Elena Simintzi look at the relationship between Wage inequality and firm growth. Rising wage inequality has received much attention recently and this column describes new evidence on the determinants of the ‘skill premium’.

There are two basic findings:

1) larger firms have grown substantially and
2) skill premia are larger at larger firms.

They therefore conclude that the growth of larger firms could help explain growing wage inequality.

To get to these results first it is necessary to identify the ‘skill premium’ and know how to measure it. The ‘skill premium’ is simply the wage difference between high and low skill workers. Defining the skill premium is one thing, measuring it is another.

Existing measures of skill premia, such as education, experience, or even occupations, are not adequate as they do not reflect a one-to-one mapping between job tasks and skill requirements. […].

In our data, provided by Income Data Services (IDS), we observe how much a firm pays workers employed in different occupations and, crucially, how these occupations map into broader ‘job level’ categories which are comparable across firms. Since job levels are determined based on the skills required for the job, comparing wages for a worker classified at a high job level to a worker classified at a low job level allows us to more directly measure the skill premium. Moreover, since we have these data for a broad cross-section of firms measured at multiple points in time, we can observe within-firm and across-time patterns in the skill premium.

To provide further detail, consider a cleaner and a finance director. The cleaner corresponds to job level 1, work that “requires basic literacy and numeracy skills and the ability to perform a few straightforward and short-term tasks to instructions under immediate supervision”. The finance director corresponds to our highest skill category – job level 9 and involves “very senior executive roles with substantial experience in, and leadership of, a specialist function, including some input to the organisation’s overall strategy”. We measure skill premium using a ratio of a high-skill to low-skill job, at the same firm, in the same year.


When examining ‘top-bottom’ wage ratios in our sample (e.g., the wage associated with job level 8 divided by the wage associated with job level 1 within the same firm and year), we find they increase with firm size. A similar, albeit weaker, relationship arises when we look at ‘top-middle’ wage ratios (e.g. the wage associated with job level 8 divided by the wage associated with job level 4 within the same firm and year). In contrast, ‘middle-bottom’ wage ratios (e.g. the wage associated with job level 4 divided by the wage associated with job level 1 within the same firm and year) stay flat, or if anything slightly decrease with firm size.

  • What is interesting is that when low job levels (1 to 5) are compared to one another, an increase in firm size has no effect on within-firm skill premia.
  • In contrast, when high job levels (6 to 9) are compared to either one another or low job levels, an increase in firm size widens the wage gap between higher and lower skill categories.

The question this give rise to is Why do wages in high-skill job categories increase with firm size but not wages in low- and medium-skill job categories?

We provide two possible explanations.

  • First, larger firms invest more in automation which allows them to replace labour with technology in certain routine jobs […].

Consistent with this hypothesis, we find that wages associated with routine jobs decline relative to those associated with non-routine jobs as firms become larger, especially in medium-skill job categories.

  • Second, larger firms may pay relatively lower entry-level managerial wages in return for providing better career opportunities […].

Consistent with this hypothesis, we find that managerial wages in low- to medium-skill job categories are relatively lower in larger firms, while those in high-skill job categories are relatively higher in larger firms.

Is there a third factor here? We know that the division of labour is limited by the extent of the market and bigger firms have larger internal labour markets which gives raise to a greater levels of specialisation with some areas of specialisation being more valuable than others. These higher value jobs receive greater remuneration.

The last question is, What do the results say about overall wage inequality?

An increasing skill premium at larger firms will lead to greater wage inequality inside those firms. But how has the size of the median employer changed over the last two decades? US firms with 500 or more employees accounted for 51.5% of all employment in 2011. As such, we measure firm size by focusing on the largest firms and find evidence of strong firm growth among larger firms in practically all of the developed countries in our sample. These results suggest that part of what may be perceived as a global trend toward more wage inequality may be driven by an increase in employment by the largest firms in the economy.

So the upshot of this is that the growth of larger firms in the economy may partially explain the rise in wage inequality seen over the last few decades.